A FIELD-TEST OF THE SOUND ENVIRONMENT HYPOTHESIS OF CONSPECIFIC SONG RECOGNITION IN AMERICAN TREE SPARROWS (SPIZELLA-ARBOREA)

The sound environment hypothesis predicts that the features of song that are important for conspecific recognition should be those that overlap least with the songs of other species in the same location. We tested this hypothesis with song playbacks to free-living male American tree sparrows (Spizella arborea). We first used a discriminant function analysis to determine which song features best separated American tree sparrow songs from the songs of heterospecific avian species at two locations representing different sound environments. Based on this analysis we predicted which song features should be most important for conspecific recognition at each location. We then produced synthetic songs containing alterations in these features. Males, however, did not respond to playbacks of altered features as predicted. Thus, our results do not support the sound environment hypothesis of conspecific song recognition.